Jump to content

  • Log In with Google      Sign In   
  • Create Account


#Actualcrybot

Posted 26 April 2013 - 08:08 AM

The data structure seems correct. The only comment I have is that if you switch to using a 16-bit representation for the move, the size of a hash entry will go from 24 bytes to 16 bytes (this depends on the compiler, of course).

 

I know, I planned to change move encoding not just for the hashentry, but also for move generation. It will take some time, so i'll do it when i can.

Anyway now a HashEntry is a 18 bytes object.


 

Can you tell me something about what i wrote in my previous post?

The last thing I have doubts about is principal variation: I'm triyng to improve move ordering and it seems that principal variation moves are the most important moves to try during alpha-beta search, but, with transposition table implemented, aren't them already the first moves tried (best moves)?


#1crybot

Posted 26 April 2013 - 07:57 AM

The data structure seems correct. The only comment I have is that if you switch to using a 16-bit representation for the move, the size of a hash entry will go from 24 bytes to 16 bytes (this depends on the compiler, of course).

 

I know, I planned to change move encoding not just for the hashentry, but also for move generation. It will take some time, so i'll do it when i can.


 

Can you tell me something about what i wrote in my previous post?

The last thing I have doubts about is principal variation: I'm triyng to improve move ordering and it seems that principal variation moves are the most important moves to try during alpha-beta search, but, with transposition table implemented, aren't them already the first moves tried (best moves)?

 

 


PARTNERS